Re: st: mice experiment, risk for a clinical manifestation

[Yoshiro Nagao <dengue_aedes@yahoo.co.jp>]

Dear Ron,

Thank you very much for your suggestion.

> So there are two outcomes:
> 1. died of Z (how irritating to die of an irrelevant
> illness...) and
> 2. died of Y
> Simplest way to handle this is in a multinomial
> logistic regression  
> with alive as the base category.

I see...I will try multinomial logistic.

By the way, mice and their immunological
conditions were monitored everyday. 
Is this a sort of survival time data (?).

In addition, mice, which died of "Z", might
have developed "Y", if they have had survived
"Z".  So, is it possible to treat 
a death from "Z" as a "failure to follow up"
in whatever sort of survival analysis?

Yoshiro

P.S. Let me apologise for the inadequate
remark about the death of mice.

--------------------------------------
Let's start Yahoo! Auction  -  Free Campaign Now!
http://pr.mail.yahoo.co.jp/auction/
*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/